U.S. patent application number 14/962286 was filed with the patent office on 2016-12-22 for wideband iq mismatch correction for direct down-conversion receiver.
The applicant listed for this patent is TEXASINSTRUMENTSINCORPORATED. Invention is credited to Jaiganesh Balakrishnan, Avinash Vasudev Sakleshpur, KARTHIK KHANNA SUBRAMANI, Nagarajan Viswanathan.
Application Number | 20160373288 14/962286 |
Document ID | / |
Family ID | 57588529 |
Filed Date | 2016-12-22 |
United States Patent
Application |
20160373288 |
Kind Code |
A1 |
SUBRAMANI; KARTHIK KHANNA ;
et al. |
December 22, 2016 |
WIDEBAND IQ MISMATCH CORRECTION FOR DIRECT DOWN-CONVERSION
RECEIVER
Abstract
A direct down-conversion (DDC) front end receiver includes first
Q-channel that filters a sum of PRBS and baseband quadrature
signals to generate a first filtered quadrature signal, a second
Q-channel that filters a difference of the baseband and PRBS
signals to generate a second filtered quadrature signal, a first
I-channel and a second I-channel, Q-path and I-path PRBS
cancellation blocks for cancelling corresponding PRBS components
from sum of first and second filtered quadrature signals and sum of
first and second filtered inphase signals respectively, Q-path and
I-path sum filter estimation blocks for estimating quadrature and
inphase sum filter responses. An IQ mismatch compensation filter
estimate and tracking block estimates IQ mismatch compensation
filter response from estimated quadrature and inphase sum filter
responses, and an IQ mismatch compensation filter filters the
modified inphase signal with the IQ mismatch compensation filter
response, to generate a filter compensated inphase signal.
Inventors: |
SUBRAMANI; KARTHIK KHANNA;
(Chennai, IN) ; Viswanathan; Nagarajan;
(Bangalore, IN) ; Sakleshpur; Avinash Vasudev;
(Bangalore, IN) ; Balakrishnan; Jaiganesh;
(Bangalore, IN) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
TEXASINSTRUMENTSINCORPORATED |
Dallas |
TX |
US |
|
|
Family ID: |
57588529 |
Appl. No.: |
14/962286 |
Filed: |
December 8, 2015 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
62182187 |
Jun 19, 2015 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
H04L 27/0014 20130101;
H04B 1/1036 20130101; H04L 27/3863 20130101; H04B 1/30
20130101 |
International
Class: |
H04L 27/38 20060101
H04L027/38; H04B 1/10 20060101 H04B001/10 |
Claims
1. A direct down-conversion front end receiver, comprising an
antenna for receiving an input radio frequency (RF) signal; first
and second mixers connected to the antenna, for mixing the input RF
signal with first and second orthogonal sinusoid signals for
generating baseband quadrature and inphase signals respectively;
first and second Q-channels connected to the first mixer, wherein
the first Q-channel adds a baseband pseudo random binary sequence
(PRBS) signal to the baseband quadrature signal, and filters the
added baseband quadrature signal with a first quadrature filter
response, to generate a first filtered quadrature signal, and
wherein the second Q-channel subtracts the baseband PRBS signal
from the baseband quadrature signal, and filters the subtracted
baseband quadrature signal with a second quadrature filter
response, to generate a second filtered quadrature signal; first
and second I-channels connected to the second mixer, wherein the
first I-channel adds the baseband PRBS signal to the baseband
inphase signal, and filters the added baseband inphase signal with
a first inphase filter response, to generate a first filtered
inphase signal, and the second I-channel subtracts the baseband
PRBS signal from the baseband inphase signal, and filters the
subtracted baseband inphase signal with a second inphase filter
response, to generate a second filtered inphase signal; Q-path and
I-path PRBS cancellation blocks, wherein the Q-path PRBS
cancellation block cancels corresponding PRBS component from a sum
of the first and the second filtered quadrature signals to generate
a modified quadrature signal, and wherein the I-path PRBS
cancellation block cancels corresponding PRBS component from a sum
of the first and the second filtered inphase signals to generate a
modified inphase signal; Q-path and I-path sum filter estimation
blocks, wherein the Q-path sum filter estimation block estimates a
quadrature sum filter response from a difference of the first and
the second filtered quadrature signals, and wherein the I-path sum
filter estimation block estimates an inphase sum filter response
from a difference of the first and the second filtered inphase
signals; an IQ mismatch compensation filter estimate and tracking
block, connected to the Q-path and I-path sum filter estimation
blocks, for estimating an IQ mismatch compensation filter response
from the estimated quadrature and the inphase sum filter responses;
and an IQ mismatch compensation filter, connected to the IQ
mismatch compensation filter estimate and tracking block, for
filtering the modified inphase signal with the IQ mismatch
compensation filter response, to generate a filter compensated
inphase signal.
2. The direct-down-conversion front end receiver as claimed in
claim 1, further comprising: a local oscillator for generating the
first orthogonal sinusoid signal of a predefined center frequency;
a phase shifter connected to the local oscillator for generating
the second orthogonal sinusoid signal orthogonal to the first
orthogonal sinusoid signal; a digital pseudo random binary sequence
(PRBS) generator for generating a PRBS signal; and a Digital to
Analog Converter (DAC) connected to the PRBS generator for
converting the generated PRBS signal to an analog baseband PRBS
signal.
3. The direct-down-conversion front end receiver as claimed in
claim 1, wherein the first and the second Q-channels include first
and second Analog to Digital converters (ADCs) for receiving the
first and the second filtered quadrature signals, and generating
one or more digital samples of the first and the second filtered
quadrature signals respectively, and wherein the first and the
second I-channels include third and fourth ADCs for receiving the
first and the second filtered inphase signals, and generating one
or more digital samples of the first and the second filtered
inphase signals.
4. The direct down-conversion front end receiver as claimed in
claim 1, further comprising a delay filter for receiving and
delaying the modified quadrature signal by a predefined delay, for
generating a filter compensated quadrature signal.
5. The direct down-conversion front end receiver as claimed in
claim 4, further comprising a mixer impairment calibration and
compensation block, for receiving and applying the mixer impairment
calibration and compensation on the filter compensated quadrature
and inphase signals, to generate mixer compensated quadrature and
inphase signals respectively.
6. The direct down-conversion front end receiver as claimed in
claim 1, wherein the quadrature sum filter response is a sum of the
first and the second quadrature filter responses, the inphase sum
filter response is a sum of the first and the second inphase filter
responses, a quadrature difference filter response is a difference
of the first and the second quadrature filter responses, and an
inphase difference filter response is a difference of the first and
the second inphase filter responses.
7. A method of IQ mismatch calibration and compensation in a
direct-down-conversion front end receiver, comprising: receiving an
input radio frequency (RF) signal; mixing the input RF signal with
first and second orthogonal sinusoid signals for generating
baseband quadrature and inphase signals respectively; adding a
baseband pseudo random binary sequence (PRBS) signal to the
baseband quadrature signal, and filtering the added baseband
quadrature signal with a first quadrature filter response to
generate a first filtered quadrature signal; subtracting the
baseband PRBS signal from the baseband quadrature signal, and
filtering the subtracted baseband quadrature signal with a second
quadrature filter response, to generate a second filtered
quadrature signal; adding the baseband PRBS signal to the baseband
inphase signal, and filtering the added baseband inphase signal
with a first inphase filter response, to generate a first filtered
inphase signal; subtracting the baseband PRBS signal from the
baseband inphase signal, and filtering the subtracted baseband
inphase signal with a second inphase filter response, to generate a
second filtered inphase signal; cancelling corresponding PRBS
component from a sum of the first and the second filtered
quadrature signals to generate a modified quadrature signal;
cancelling corresponding PRBS component from a sum of the first and
the second filtered inphase signals to generate a modified inphase
signal; estimating a quadrature sum filter response based on a
difference of the first and the second filtered quadrature signals;
estimating an inphase sum filter response based on a difference of
the first and the second filtered inphase signals; estimating an IQ
mismatch compensation filter response based on the estimated
quadrature and the inphase sum filter responses; and filtering the
modified inphase signal with the IQ mismatch compensation filter
response, to generate a filter compensated inphase signal.
8. The method as claimed in claim 7, further comprising delaying
the modified quadrature signal by a predefined delay, for
generating a filter compensated quadrature signal.
9. The method as claimed in claim 7, further comprising applying
mixer impairment calibration and a compensation on the filter
compensated quadrature and inphase signals, to generate mixer
compensated quadrature and inphase signals respectively.
10. The method as claimed in claim 7, wherein the quadrature sum
filter response is a sum of the first and the second quadrature
filter responses, the inphase sum filter response is a sum of the
first and the second inphase filter responses, a quadrature
difference filter response is a difference of the first and the
second quadrature filter responses, and an inphase difference
filter response is a difference of the first and the second inphase
filter responses.
11. A direct down-conversion front end receiver, comprising: an
antenna for receiving an input radio frequency (RF) signal; a
signal adder, connected to the antenna, for adding a pseudo random
binary sequence (PRBS) signal of Radio frequency (RF), to the input
RF signal to generate a modified input RF signal; first and second
mixers connected to the signal adder, for mixing the modified input
RF signal with first and second orthogonal sinusoid signals for
generating baseband quadrature and inphase signals respectively;
first and second low pass filters (LPFs) connected to the first and
second mixers respectively, for filtering the baseband quadrature
and inphase signals with quadrature and inphase filter responses,
for generating filtered quadrature and inphase signals
respectively; first and second Analog to Digital Converter (ADCs)
connected to the first and second low pass filters (LPFs), for
converting the filtered quadrature and inphase signals to a digital
form; an adder connected to the first and the second ADCs for
receiving and adding the digitized filtered quadrature and inphase
signals, to generate an overall down-converted complex baseband
signal, wherein the overall down-converted complex baseband signal
includes baseband input and PRBS signals filtered by a signal
transfer function, and the baseband input and PRBS image signals
filtered by an image transfer function; an IQ mismatch calibration
block for estimating the signal transfer and the image transfer
functions affecting the baseband PRBS signal in the presence of the
overall down-converted baseband signal, based on the overall
down-converted complex baseband signal and the baseband PRBS
signal; an IQ mismatch compensation block for generating an 1Q
compensated baseband signal based on the overall down-converted
complex baseband signal and the estimated signal transfer and the
image transfer functions; and a PRBS cancellation block for
cancelling PRBS components from the IQ compensated baseband signal
based on the baseband PRBS signal, and the estimated signal
transfer and the image transfer functions, to generate an IQ
compensated baseband input signal.
12. The direct down-conversion front end receiver as claimed in
claim 11, further comprising: a baseband PRBS signal generator for
generating the baseband PRBS signal; and a super-heterodyne
upconverter, connected to the baseband PRBS signal generator, for
receiving and up-converting the baseband PRBS signal to a radio
frequency (RF) PRBS signal.
13. The direct down-conversion front end receiver as claimed in
claim 11, wherein the IQ mismatch calibration block comprises: a
PRBS filter based signal transfer function estimation block for
estimating a signal transfer function associated with the baseband
PRBS signal in the presence of the overall down-converted complex
baseband signal; and a PRBS filter based image transfer function
estimation block for estimating an image transfer function
associated with the baseband PRBS image signal in the presence of
the overall down-converted complex baseband signal.
14. The direct down-conversion front end receiver as claimed in
claim 11, wherein the IQ compensated baseband signal
r.sub.bb,comp(n) is represented by the expression:
r.sub.bb,comp(n)=k.sub.sig*(n)*r.sub.bb(n)-k.sub.img(n)*r.sub.bb*(n)
where r.sub.bb(n) is the overall down-converted complex baseband
signal, r.sub.bb*(n) is an overall down-converted baseband image
signal, k.sub.sig(n) and k.sub.img(n) are signal transfer and image
transfer functions respectively, and
r.sub.bb(n)=k.sub.sig(n)*s.sub.bb(n)+k.sub.img(n)*s.sub.bb*(n)+k.sub.sig(-
n)*p.sub.bb(n)+k.sub.img(n)*p.sub.bb*(n) where s.sub.bb(n) is
baseband input signal, s.sub.bb*(n) is baseband input image signal,
p.sub.bb(n) is baseband PRBS signal, and p.sub.bb*(n) is baseband
PRBS image signal.
15. A method of IQ mismatch calibration and compensation in a
direct down-conversion front end receiver, comprising: receiving an
input radio frequency (RF) signal; adding a pseudo random binary
sequence (PRBS) signal of Radio frequency (RF), to the input RF
signal to generate a modified input RF signal; mixing the modified
input RF signal with first and second orthogonal sinusoid signals
for generating baseband quadrature and inphase signals
respectively; filtering the baseband quadrature and inphase signals
with quadrature and inphase filter responses, for generating
filtered quadrature and inphase signals respectively; converting
the filtered quadrature and inphase signals to a digital form;
adding the digitized filtered quadrature and inphase signals, to
generate an overall down-converted complex baseband signal, wherein
the overall down-converted complex baseband signal includes
baseband input and PRBS signals filtered by a signal transfer
function, and the baseband input and PRBS image signals filtered by
an image transfer function; estimating the signal transfer and
image transfer functions affecting the baseband PRBS signal in
presence of the overall down-converted complex baseband signal,
based on the overall down-converted complex baseband signal and the
baseband PRBS signal; generating an IQ compensated baseband signal
based on the overall down-converted complex baseband signal and the
estimated signal transfer and the image transfer functions; and
cancelling PRBS components from the IQ compensated baseband signal
based on the baseband PRBS signal, and the estimated signal
transfer and the image transfer functions, to generate an IQ
compensated baseband input signal.
16. The method as claimed in claim 15, further comprising:
generating a baseband PRBS signal; and filtering with a zero order
hold image rejection filter and up-converting the baseband PRBS
signal to a radio frequency (RF) PRBS signal.
17. The method as claimed in claim 15, wherein the IQ compensated
baseband signal r.sub.bb,comp(n) is represented by the expression:
r.sub.bb,comp(n)=k.sub.sig*(n)*r.sub.bb(n)-k.sub.img(n)*r.sub.bb*(n)
where r.sub.bb(n) is the overall down-converted baseband signal,
r.sub.bb*(n) is an overall down-converted baseband image signal,
k.sub.sig(n) and k.sub.img(n) are the signal transfer and the image
transfer functions respectively, and
r.sub.bb(n)=k.sub.sig(n)*s.sub.bb(n)+k.sub.img(n)*s.sub.bb*(n)+k.sub.sig(-
n)*p.sub.bb(n)+k.sub.img(n)*p.sub.bb*(n) where s.sub.bb(n) is
baseband input signal, s.sub.bb*(n) is baseband input image signal,
p.sub.bb(n) is baseband PRBS signal, and p.sub.bb*(n) is baseband
PRBS image signal.
Description
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] This application claims the benefit of U.S. Provisional
Patent Application No. 62/182,187, filed Jun. 19, 2015 and that is
incorporated herein by reference.
TECHNICAL FIELD
[0002] The present disclosure generally relates to IQ mismatch
correction in direct down-conversion receivers and super heterodyne
receivers, for example, IQ mismatch correction in multicarrier GSM
base station receivers.
BACKGROUND
[0003] FIG. 1 illustrates a conventional direct-down-conversion
(DDC) front end receiver 100. At mixers 102a and 102b, an input
Radio frequency (RF) signal r(t) is mixed with two orthogonal
sinusoids, 2gIcos(2.pi.Fct+.PHI.I) and -2gQsin(2.pi.Fct+.PHI.K), to
generate inphase (Im) and Quadrature (Qm) components respectively,
where gI & gQ are amplitudes of sinusoids and .PHI.I &
.PHI.Q are phases of sinusoids. Ideally, gI=gQ for equality of
sinusoid amplitude and .PHI.1=.PHI.Q for sinusoid orthogonality.
However, and practically, gI.noteq.gQ and .PHI.I.noteq..PHI.Q.
[0004] The mixed down baseband (BB) signal (rm.sub.bb(t)) generated
after mixing is given by rm.sub.bb(t)=I.sub.m(t)+jQ.sub.m(t), and
is represented by the following expression:
rm.sub.bb(t)=K.sub.sigs.sub.bb(t)+K.sub.imgs.sub.bb*(t) (1)
Where,
[0005] s.sub.bb(t)=desired signal
K sig = ( 1 + g - j.phi. 2 ) K img = ( 1 - g j.phi. 2 )
##EQU00001## g ( relative gain imbalance ) = gQ / gI ##EQU00001.2##
.PHI. ( relative phase imbalance ) = .PHI. Q = .PHI. I
##EQU00001.3##
[0006] As seen in equation (1), apart from the desired signal
(s.sub.bb(t))), a scaled version of an undesirable image signal
(s.sub.bb(t)) also appears due to mixer gain and phase impairments.
The magnitude of the image signal depends on the relative phase and
gain imbalances.
[0007] The Inphase (I.sub.m) and Quadrature (Q.sub.m) components of
the mixed down baseband (BB) signal (rm.sub.bb(t)) pass through
first and second low pass filters (104a and 104b respectively) to
generate filtered Inphase and Quadrature (I.sub.f) and (Q.sub.f)
components respectively. The overall baseband signal Rf.sub.bb(f)
(in frequency domain) after filtering is given by
Rf.sub.bb(f)=I.sub.f(f)+jQ.sub.f(f)=I.sub.m(f)HI(f)+jQ.sub.m(f)HQ(f),
and is represented by the following expression:
Rf bb ( f ) = K sig ( f ) S bb ( f ) + K img ( f ) S bb * ( - f ) (
2 ) Where , K sig ( f ) = ( 1 + HQ ( f ) HI ( f ) g - j.phi. 2 ) HI
( f ) and K img ( f ) = ( 1 - HQ * ( - f ) HI * ( - f ) g j.phi. 2
) HI * ( - f ) = ( 1 - HQ ( f ) HI ( f ) g j.phi. 2 ) HI ( f ) HI (
f ) = impulse response of the first filter 104 a HQ ( f ) = impulse
response of the second filter 104 b H ( f ) = ( relative filter
imbalance ) = HQ ( f ) / HI ( f ) ( 3 ) ##EQU00002##
[0008] As seen in equation (2), apart from a scaled (K.sub.sig(f))
version of the desired signal (S.sub.bb(f)), a scaled
(K.sub.img(f)) version of the image signal (S*.sub.bb(f)) also
appears due to the relative filter imbalance. The magnitude of the
image signal (S*.sub.bb(-f)) depends on the relative filter
imbalance H(f)=HQ(f)/HI(f).
[0009] The mismatch between the inphase and quadrature components
causes image signals, making the use of direct down-conversion
unfeasible for multicarrier receivers. It is desirable to calibrate
IQ mismatch in baseband receivers with an image rejection ratio
(IRR) greater than 90 dB. Further, the IQ mismatch calibration
should be done in background, as separate calibration duration
cannot be availed. Furthermore, the IQ mismatch calibration time
should be less than 500 ms and should not significantly affect the
GSM BS boot up time.
SUMMARY
[0010] This Summary is provided to introduce a selection of
concepts in a simplified form that are further described below in
the Detailed Description. This Summary is not intended to identify
key or important features of the claimed subject matter, nor is it
intended to be used as an aid in determining the scope of the
claimed subject matter.
[0011] A direct down-conversion front end receiver, a direct
down-conversion front end receiver, and one or more methods of IQ
mismatch correction and calibration are disclosed. In an
embodiment, a direct down-conversion front end receiver is
disclosed that includes an antenna for receiving an input radio
frequency (RF) signal and first and second mixers connected to the
antenna, for mixing the input RF signal with first and second
orthogonal sinusoid signals for generating baseband quadrature and
inphase signals respectively. First and second Q-channels are
connected to the first mixer, wherein the first Q-channel adds a
baseband pseudo random binary sequence (PRBS) signal to the
baseband quadrature signal, and filters the added baseband
quadrature signal with a first quadrature filter response, to
generate a first filtered quadrature signal, and wherein the second
Q-channel subtracts the baseband PRBS signal from the baseband
quadrature signal, and filters the subtracted baseband quadrature
signal with a second quadrature filter response, to generate a
second filtered quadrature signal. Further, first and second
I-channels are connected to the second mixer, wherein the first
I-channel adds the baseband PRBS signal to the baseband inphase
signal, and filters the added baseband inphase signal with a first
inphase filter response, to generate a first filtered inphase
signal, and the second I-channel subtracts the baseband PRBS signal
from the baseband inphase signal, and filters the subtracted
baseband inphase signal with a second inphase filter response, to
generate a second filtered inphase signal. The direct
down-conversion front end receiver further includes Q-path and
I-path PRBS cancellation blocks, wherein the Q-path PRBS
cancellation block cancels corresponding PRBS component from a sum
of the first and the second filtered quadrature signals to generate
a modified quadrature signal, and wherein the I-path PRBS
cancellation block cancels corresponding PRBS component from a sum
of the first and the second filtered inphase signals to generate a
modified inphase signal. The direct down-conversion front end
receiver furthermore includes Q-path and I-path sum filter
estimation blocks, wherein the Q-path sum filter estimation block
estimates a quadrature sum filter response from a difference of the
first and the second filtered quadrature signals, and wherein the
I-path sum filter estimation block estimates an inphase sum filter
response from a difference of the first and the second filtered
inphase signals. An IQ mismatch compensation filter estimate and
tracking block are connected to the Q-path and I-path sum filter
estimation blocks, for estimating an IQ mismatch compensation
filter response from the estimated quadrature and the inphase sum
filter responses. Further, an IQ mismatch compensation filter is
connected for filtering the modified inphase signal with the IQ
mismatch compensation filter response, to generate a filter
compensated inphase signal.
[0012] Additionally, in an embodiment, a method of IQ mismatch
calibration and compensation in a direct down-conversion front end
receiver is disclosed. The method includes receiving an input radio
frequency (RF) signal, and mixing the input RF signal with first
and second orthogonal sinusoid signals for generating baseband
quadrature and inphase signals respectively. The method further
includes adding a baseband pseudo random binary sequence (PRBS)
signal to the baseband quadrature signal, filtering the added
baseband quadrature signal with a first quadrature filter response
to generate a first filtered quadrature signal, subtracting the
baseband PRBS signal from the baseband quadrature signal, filtering
the subtracted baseband quadrature signal with a second quadrature
filter response, to generate a second filtered quadrature signal,
adding the baseband PRBS signal to the baseband inphase signal,
filtering the added baseband inphase signal with a first inphase
filter response, to generate a first filtered inphase signal,
subtracting the baseband PRBS signal from the baseband inphase
signal, and filtering the subtracted baseband inphase signal with a
second inphase filter response, to generate a second filtered
inphase signal. The method furthermore includes cancelling
corresponding PRBS component from a sum of the first and the second
filtered quadrature signals to generate a modified quadrature
signal, and cancelling corresponding PRBS component from a sum of
first and second filtered inphase signals to generate a modified
inphase signal. The method further includes estimating a quadrature
sum filter response based on a difference of the first and the
second filtered quadrature signals, estimating an inphase sum
filter response based on a difference of the first and second
filtered inphase signals, estimating an IQ mismatch compensation
filter response based on the estimated quadrature and the inphase
sum filter responses, and filtering the modified inphase signal
with the IQ mismatch compensation filter response, to generate a
filter compensated inphase signal.
[0013] Moreover, in an embodiment, a direct down-conversion front
end receiver is disclosed, that includes an antenna for receiving
an input radio frequency (RF) signal, a signal adder, connected to
the antenna, for adding a pseudo random binary sequence (PRBS)
signal of Radio frequency (RF), to the input RF signal to generate
a modified input RF signal. First and second mixers are connected
to an adder, for mixing the modified input RF signal with first and
second orthogonal sinusoid signals for generating baseband
quadrature and inphase signals respectively. First and second low
pass filters (LPFs) are connected to the first and second mixers
respectively, for filtering the baseband quadrature and inphase
signals with quadrature and inphase filter responses, for
generating filtered quadrature and inphase signals respectively.
First and second Analog to Digital Converters (ADCs) are connected
to the first and second low pass filters (LPFs), for converting the
filtered quadrature and inphase signals to a digital form. An adder
is connected to the first and the second ADCs for receiving and
adding the digitized filtered quadrature and inphase signals, to
generate an overall down-converted complex baseband signal, wherein
the overall down-converted baseband signal includes baseband input
and PRBS signals filtered by a signal transfer function, and the
baseband input and PRBS image signals filtered by an image transfer
function. The direct down-conversion front end receiver includes an
IQ mismatch calibration block for receiving the overall
down-converted baseband signal and the baseband PRBS signal, and
estimating the signal transfer and the image transfer functions
affecting the baseband PRBS signal in the presence of the overall
down-converted baseband signal, an IQ mismatch compensation block
for receiving the overall down-converted baseband signal and the
estimated signal transfer and image functions, and generating an IQ
compensated baseband signal. The direct down-conversion front end
receiver furthermore includes a PRBS cancellation block for
receiving the IQ compensated baseband signal, baseband PRBS signal,
and the estimated signal transfer and the image transfer functions,
and cancelling PRBS components from the IQ compensated baseband
signal and the estimated signal transfer and the image transfer
functions, to generate an IQ compensated baseband input signal free
of PRBS components.
[0014] Other and additional aspects and example embodiments are
provided in the drawings and the detailed description that
follow.
BRIEF DESCRIPTION OF THE FIGURES
[0015] FIG. 1 is a schematic block diagram of a conventional
direct-down-conversion (DDC) front end receiver;
[0016] FIG. 2 is a schematic block diagram of a DDC front end
receiver, in accordance with an embodiment of the present
disclosure;
[0017] FIG. 3 is a schematic block diagram of a PRBS based filter
estimation block for estimating a PRBS based filter response, in
accordance with an embodiment of the present disclosure;
[0018] FIG. 4 is a schematic block diagram of a PRBS cancellation
block, in accordance with an embodiment of the present
disclosure;
[0019] FIGS. 5A and 5B are a flow diagram chart of a method of IQ
mismatch calibration and compensation in the DDC front end
receiver, in accordance with an embodiment of the present
disclosure;
[0020] FIG. 6 is a schematic block diagram of a DDC front end
receiver, in accordance with an embodiment of the present
disclosure;
[0021] FIG. 7 is a schematic block diagram of a signal generation
block for generating a PRBS signal of RF, in accordance with an
embodiment of the present disclosure; and
[0022] FIG. 8 is a flow diagram chart of a method of IQ mismatch
calibration and compensation in the DDC front end receiver, in
accordance with an embodiment of the present disclosure.
[0023] The drawings referred to in this description are not to be
understood as being drawn to scale except if specifically noted,
and such drawings are only exemplary in nature.
DETAILED DESCRIPTION
[0024] Referring to FIG. 2, a schematic block diagram of a dual
channel direct-down-conversion (DDC) front end receiver 200, in
accordance with an embodiment of the present disclosure is
illustrated. The dual channel DDC front end receiver 200 includes
an antenna 201 for receiving an input radio frequency (RF) signal
r(t), first and second mixers 202a and 202b for mixing the input RF
signal r(t) with first and second orthogonal sinusoid signals, for
generating baseband Quadrature (Qm) and Inphase (Im) signals in Q
and I paths respectively, a local oscillator (LO) 204 for
generating the first sinusoid signal 2gIcos(2.pi.Fct+.PHI.I) of
centre frequency F.sub.c, and a 90 degree phase shifter 205,
connected to the LO 204 for generating the second sinusoid signal
-2gQsin(2.pi.F.sub.ct+.PHI.Q) orthogonal to the first sinusoid
signal. The gI and gQ are amplitudes of the first and second
sinusoid signals respectively, and .PHI.I and .PHI.Q are phases of
the first and second sinusoid signals respectively. The unequal
values of gI and gQ, and unequal values of .PHI.I and .PHI.Q
introduces mismatch between baseband Quadrature (Qm) and Inphase
(Im) signals.
[0025] The dual channel DDC front end receiver 200 further includes
a digital pseudo random binary sequence (PRBS) generator 206 for
generating a PRBS signal, and a Digital to Analog Converter (DAC)
208 connected to the PRBS generator, for converting the generated
PRBS signal to an analog baseband PRBS signal P.
[0026] In the dual channel DDC front end receiver 200, each of the
Q and I paths includes two channels, for example, the Q-path
includes first and second Q-channels 210a and 210b, and the I-path
includes first and second I-channels 211a and 211b.
[0027] In the first Q-channel 210a, a first adder 212a adds the
baseband Quadrature signal Qm to the baseband PRBS signal P to
generate an added Quadrature signal Qm1, represented by the
following expression:
Qm1=Qm+P (4)
[0028] The added Quadrature signal Qm1 passes through a first LPF
214a (of a first quadrature filter response HQ1(f)) to generate a
first filtered Quadrature signal Qf1, represented by the following
expression:
Qf1=Qm1HQ1(f)=(Qm+P)HQ1(f) (5)
[0029] In the second Q-channel 210b, a first subtractor 213a
subtracts the baseband PRBS signal P from the baseband Quadrature
signal Qm to generate a subtracted Quadrature signal Qm2,
represented by the following expression:
Qm2=Qm-P (6)
[0030] The subtracted Quadrature signal Qm2 passes through a second
LPF 214b (of a second quadrature filter response HQ2(f)) to
generate a second filtered Quadrature signal Qf2, represented by
the following expression:
Qf2=Qm2HQ2(f)=(Qm-P)HQ2(f) (7)
[0031] First and second ADCs 216a and 216b generate digital samples
of the first and second filtered Quadrature signals Qf1 and Qf2 for
further processing.
[0032] Similarly, in the first I-channel 211a, a second adder 212b
adds the baseband Inphase component Im to the baseband PRBS signal
P to generate an added Inphase signal Im1, represented by the
following expression:
Im1=Im+P (8)
[0033] The added Inphase signal Im1 passes through a third LPF 214c
(of a first inphase filter response (HI1(f)) to generate a first
filtered Inphase signal If1, represented by the following
expression:
If1=Im1HI1(f)=(Im+P)HI1(f) (9)
[0034] In the second I-channel 211b, a second subtractor 213b
subtracts the baseband Inphase component Im from the baseband PRBS
signal P to generate a subtracted Inphase signal Im2, represented
by the following expression:
Im2=Im-P (10)
[0035] The difference Inphase signal Im2 passes through a fourth
LPF 214d (of a second inphase filter response (HI2(f)) to generate
a second filtered Inphase signal If2, represented by the following
expression:
If2=Im1HI2(f)=(Im-P)HI2(f) (11)
[0036] Third and fourth ADCs 216c and 216d generate digital samples
of the first and second filtered Inphase signals If1 and If2 for
further processing.
[0037] Further, in the dual channel DDC front end receiver 200, a
third adder 212c adds the first and second filtered Quadrature
signals Qf1 and Qf2 to generate a sum quadrature signal Qs,
represented by the following expression:
Qs=Qf1+Qf2=QmHQs+PHQd (12)
[0038] A third subtractor 213c generates a difference of the first
and second filtered Quadrature signals Qf1 and Qf2 to generate a
difference quadrature signal Qd, represented by the following
expression:
Qd=Qf1-Qf2=QmHQd+PHQs (13)
Where,
HQs (quadrature sum filter response)=HQ1(f)+HQ2(f)
HQd (quadrature difference filter response)=HQ1(f)-HQ2(f)
[0039] The quadrature sum filter response HQs is the overall
desired filter response in the Q-path, and Qm, HQs is desired
filtered quadrature signal in the Q-path. The quadrature difference
filter response HQd is undesired filter response, generally small
in magnitude, i.e. less than 40 dB, as HQ1(f) and HQ2(f) have
approximately similar values, and PHQd is undesired differential
filtered PRBS signal, that needs to be cancelled from the sum
Quadrature signal Qs.
[0040] Similarly, a fourth adder 212d generates a sum Inphase
signal Is of the first and second filtered Inphase signals If1 and
If2, which is represented by the following expression:
Is=If1+If2=ImHIs+PHId (14)
[0041] A fourth subtractor 213d generates a difference Inphase
signal Id from the first and second filtered Inphase signals If1
and If2, which is represented by the following expression:
Id=If1-If2=ImHId+PHIs (15)
Where,
HIs (Inphase sum filter response)=HI1(f)+HI2(f)
HId (Inphase difference filter response)=HI1(f)-HI2(f)
[0042] The inphase sum filter response HIs is the overall desired
filter response in the I-path, and ImHIs is desired filtered
inphase signal in the I-path. The inphase difference filter
response HId is an undesired filter response, and is generally
small in magnitude, i.e. less than 40 dB, as HI1(f) and HI2(f) have
approximately similar values, and PHId is undesired differential
filtered PRBS signal, that needs to be cancelled from the sum
Inphase signal Is.
[0043] The dual channel DDC front end receiver 200 further includes
Q-path and I-path PRBS cancellation blocks 220a and 220b for
receiving sum Quadrature and Inphase signals Qs and Is, and
cancelling corresponding differential filtered PRBS components PHQd
and PHId therefrom, to generate modified quadrature and inphase
signals Qsc and Isc that are free of PRBS signals, such that,
Qsc=Qs-PHQd=QmHQs (16)
Isc=Is-PHId=ImHIs (17)
[0044] The dual channel DDC front end receiver 200 further includes
Q-path and I-path sum filter estimation blocks 222a and 222b, for
receiving difference Quadrature and Inphase signals Qd and Id
respectively, and estimating quadrature and inphase sum filter
responses HQs and HIs associated with sum filtered PRBS components
PHQs and PHIs, based on a PRBS based filter estimation technique
explained in detail with reference to FIG. 3.
[0045] There is a mismatch between quadrature and inphase sum
filter responses HQs and HIs, due to filter mismatch in first
through fourth filters 214a till 214d. In order to compensate the
mismatch due to filter impairment, an IQ mismatch compensation
filter 224 (having filter response HIc(f)=HQs(f)/HIs(f)) is
provided in the I-path and an appropriate delay filter 226 is
provided in the Q-path to compensate the filter mismatch.
[0046] The dual channel DDC front end receiver 200 further includes
an IQ mismatch compensation filter estimate block 228, for
receiving quadrature and inphase sum filter responses HQs and HIs,
and generating an IQ mismatch compensation filter response HIc(f),
represented by the following expression:
HIc(f)=HQs(f)/HIs(f) (18)
[0047] In an example scenario, the IQ mismatch compensation filter
estimate and tracking block 228 estimates the compensation filter
response HIc(f) from HIs(f) and HQs(f), such that
HQs(f)=HIs(f)*HIc(f) using existing "Recursive Deconvolution"
techniques.
[0048] The IQ mismatch compensation filter 224 generates a filter
compensated Inphase signal Ifc based on the IQ mismatch
compensation filter response HIc(f) estimated by the IQ mismatch
compensation filter estimate and tracking block 228, where Ifc is
represented by the following expression:
Ifc=ImHIsHIc=ImHIs(HQs/HIs)=ImHQs (19)
[0049] Alongside, the delay filter 226 generates a filter
compensated Quadrature signal Qfc, represented by the following
expression:
Qfc=QmHQs (20)
[0050] As seen in equations 18 and 19, both the filter compensated
Quadrature and inphase signals Qfc and Ifc have identical filter
response component HQs, and therefore there is approximately no
filter mismatch between the filter compensated Quadrature and
inphase signals Qfc and Ifc.
[0051] The dual channel DDC front end receiver 200 further includes
a mixer impairment calibration and compensation block 230, for
receiving filter compensated quadrature and inphase Qfc and Ifc
signals, and applying mixer impairment calibration and compensation
thereon, to generate mixer compensated Quadrature and Inphase Qmc
and Imc signals respectively. The filter compensated quadrature and
inphase Qfc and Ifc signals have theoretically only the mixer
impairment, which can be calibrated and compensated using well
known mixer impairment and calibration techniques.
[0052] FIG. 3 is a schematic block diagram of a PRBS based filter
estimation block 300 for estimating PRBS based filter response, in
accordance with an embodiment of the present disclosure. The PRBS
based filter estimation block 300 estimates a filter response
function (hp(n)) affecting a PRBS signal p(n) in the presence of a
strong signal x(n). An input signal y(n) to the PRBS based filter
estimation block 300 is sum of an input signal x(n) and a PRBS
signal p(n) filtered by a filter response hp(n) of length Lp. The
input signal y(n) can be represented (in time domain) as below
where `*` is the time domain convolution operation:
y ( n ) = x ( n ) + p ( n ) * hp ( n ) = x ( n ) + k = 0 Lp - 1 p (
n - k ) hp ( k ) ( 21 ) ##EQU00003##
[0053] In an embodiment of the present disclosure, the PRBS based
filter estimation block 300 is employed in Q and I paths sum filter
estimation blocks 222a and 222b, for determining sum filter
responses HQs and HIs affecting PRBS signal P in presence of strong
signals QmHQd and ImHId respectively, where the input signals y(n)
are difference quadrature and inphase signals Qd and Id
respectively (See equations 12 and 14).
[0054] Referring back to FIG. 3, the PRBS based filter estimation
block 300 includes first through L-1 delay elements 302a till 302k
for generating L-1 delayed samples of the conjugate PRBS signal
p*(n), first through L multipliers 304a till 304m for multiplying
the input signal y(n) with L number of samples of the conjugate
PRBS signal p*(n). The multiplication of the input signal y(n) with
p*(n-m) (where * is the complex conjugate operation) for
0.ltoreq.m.ltoreq.Lp-1 is represented by the following
expression:
y ( n ) p * ( n - m ) = x ( n ) p * ( n - m ) + k = 0 Lp - 1 p ( n
- k ) hp ( k ) p * ( n - m ) y ( n ) p * ( n - m ) = hp ( m ) p ( n
- m ) 2 + k = 0 , k = m Lp - 1 p ( n - k ) hp ( k ) p * ( n - m ) +
x ( n ) p * ( n - m ) ( 22 ) ##EQU00004##
[0055] In the equation 21, apart from filter tap coefficient of
hp(m), some undesired terms (second term onwards) are also present.
By proper choice of PRBS sequence and averaging over large number
of samples, the second term can be set as very small or zero. Thus,
m'th coefficient of hp(n) is given by following operation with
proper scaling for unity PRBS power
hp(m)=En[y(n)p*(n-m)] (23)
where En[.] is the expectation operation over sample index `n`. The
PRBS based filter estimation block 300 includes first through L
expectation operation blocks 306a till 306m for generating L
coefficients of hp(n).
[0056] FIG. 4 is a schematic block diagram of a PRBS cancellation
block 400, which is an example of the I-path PRBS cancellation
block 220b, for cancelling differential filtered PRBS component
PHId from the sum inphase signal Is, in accordance with an
embodiment of the present disclosure. The PRBS cancellation block
400 includes a PRBS based filter estimation block 402, a filter
coefficient tracking block 404, a filter 406, and a subtractor
408.
[0057] The PRBS based filter estimation block 402 (similar to the
PRBS based filter estimation block 300) receives the sum Inphase
signal Is and the baseband PRBS signal P, and generates an estimate
of difference filter response HId(est) affecting the PRBS signal P.
The filter coefficient tracking block 404 tracks the estimate of
difference filter response HId(est) to generate a tracked value of
difference filter response HId, i.e. HId(track). The filter 406
filters the PRBS signal P with a tracked value of HId, i.e. HId
(track) to generate a differential filtered PRBS component
PHId(track). The subtractor 408 subtracts the differential filtered
PRBS component PHId(track) from the sum inphase signal Is to
generate a modified inphase signal Isc that is free of PRBS signal.
After the PRBS cancellation, the resulting modified inphase signal
Isc is equivalent to mixed down inphase signal Im filtered by the
sum filter response HIs(which is similar in characteristics to an
individual channel filter). Although not illustrated separately, it
would be apparent to one of ordinary skill in the art that the PRBS
cancellation block 400 can be used in a similar manner for
cancelling differential filtered PRBS component PHQd component from
the sum Quadrature signal Qs to generate a modified Quadrature
signal Qsc that is free of PRBS signal P.
[0058] In an example scenario, the PRBS cancellation error
P*(HId-HId(track)) and P*(HQd-HQd(track)) in the I and Q paths is
below -90 dBfs to meet the overall noise floor requirement and an
image rejection ratio requirement of being greater than 90 dB. In
an example scenario, the PRBS cancellation error is due to filter
estimation error, when may occur due to a slight mismatch between
differential filter response and corresponding tracked filter
response. An IRR of 96 dB for filter mismatch and Signal to
residual PRBS power ratio (SPR) of 96 dB is targeted so that
overall signal (image+residual PRBS) power ratio of 93 dB is
maintained. In order to achieve a filter mismatch IRR of 96 dB, the
sum filter responses, HIs and HQs, are estimated to an accuracy
.gtoreq.96 dB. In order to achieve a SPR of 96 dB, the difference
filter responses HId and HQd are estimated to an accuracy
.gtoreq.96 dB.
[0059] Further, while estimation of inphase sum filter response HIs
from (Id=ImHId+PHIs) and quadrature sum filter response HQs from
(Qd=QmHQd+PHQs), for a signal (ImHId)/(QmHQd) of power [Px] dBFS,
PRBS signal P of power [Pp] dBFS, the average processing gain ([PG]
dB) required for a filter estimation error [Err] dBc w.r.t 0 dBFS
is given by following expression:
[Px]-[PG].ltoreq.[Pp]+[Err][PG].gtoreq.[Px]-[Pp]-[Err] (24)
[0060] The minimum number of signal samples NSampPG (generated by
the each of the ADCs 216a-216d) for achieving the above-mentioned
processing gain ([PG] dB) in each of the Q and I paths is given by
the following expression:
NSampPG .gtoreq. 10 [ PG ] 10 ( 25 ) ##EQU00005##
[0061] In an example scenario, when the inphase and quadrature
difference filter responses HId and HQd has a signal power level of
-40 dBc (i.e) |HId|.sup.2=-40 dBc and |HQd|.sup.2.ltoreq.-40 dBc,
and Inphase and Quadrature Im and Qm signals has a signal power
level of 0 dBFS, the effective signal power level [Px] of each of
(ImHId) and (QmHQd) is equal to -40 dBFS.
[0062] Further, when the baseband PRBS signal P has a signal power
level [Pp]=-20 dBFS, and a required filter estimation error
[Eerr].ltoreq.--90 dBc, then as per equation 23, the required
Processing Gain (PG) in each of the Q and I paths is estimated to
be greater than or equal to 70 dBc.
[0063] As per equation 24, the minimum number of ADC samples
(NSampPG) required for a processing gain of 70 dBc is estimated to
be 10.sup.7, which is a drastic reduction as compared to 10.sup.11
samples required in existing single channel PRBS based
architecture. At a sampling rate 250 MSPS of ADC, this translates
to a calibration time of 40 ms, which is preferable as compared to
400 s in a single channel architecture. Thus, due to low signal
power level [Px] of (ImHId) and (QmHQd), the calibration time is
reduced, where the low signal power level [Px] of (ImHId) and
(QmHQd) is due to presence of low power differential quadrature and
inphase filter responses HId and HQd, which in turn is due to
suppressing of high power signals on subtraction of filtered
signals.
[0064] FIGS. 5A and 5B are a flow diagram chart of a method 500 of
IQ mismatch calibration and compensation in the dual channel DDC
front end receiver, in accordance with an embodiment of the present
disclosure. In certain embodiments, operations of the method 500
are performed in the dual channel DDC front end receiver, such as,
for example, the DDC front end receiver 200 (see, e.g., FIG.
2).
[0065] At block 502, the method includes receiving an input radio
frequency (RF) signal r(t).
[0066] At block 504, the method includes mixing the input RF signal
r(t) with first and second orthogonal sinusoid signals
2gIcos(2.pi.Fct+.PHI.I) and -2gQsin(2.pi.F.sub.ct+.PHI.Q), for
generating baseband inphase (Im) and quadrature (Qm) signals
respectively.
[0067] At block 506, the method includes adding a baseband pseudo
random binary sequence (PRBS) signal P to the baseband quadrature
signal Qm, and filtering the added baseband quadrature signal
(Qm+P) with a first quadrature filter response (HQ1(f)) to generate
a first filtered quadrature signal Qf1.
[0068] At block 508, the method includes subtracting the baseband
PRBS signal P from the baseband quadrature signal Qm, and filtering
the subtracted baseband quadrature signal (Qm-P) with a second
quadrature filter response (HQ2(f)), to generate a second filtered
quadrature signal Qf2.
[0069] At block 510, the method includes adding the baseband PRBS
signal P to the baseband inphase signal Im, and filtering the added
baseband Inphase signal (Im+P) with a first inphase filter response
(HI1(f)), to generate a first filtered inphase signal If1.
[0070] At block 512, the method includes subtracting the baseband
PRBS signal P from the baseband inphase signal Im, and filtering
the subtracted baseband inphase signal (Im-P) with a second inphase
filter response (HI2(f)), to generate a second filtered inphase
signal If2.
[0071] At block 514, the method includes adding the first and
second filtered quadrature signals Qf1 and Qf2, to generate a sum
quadrature signal Qs.
[0072] At block 516, the method includes adding the first and
second filtered inphase signals If1 and If2, to generate a sum
inphase signal Is.
[0073] At block 518, the method includes subtracting the second
filtered quadrature signal Qf2 from the first filtered quadrature
signal Qf1, to generate a difference quadrature signal Qd.
[0074] At block 520, the method includes subtracting the second
filtered inphase signal If2 from the first filtered inphase signal
If1, to generate a difference inphase signal Id.
[0075] At block 522, the method includes cancelling corresponding
PRBS signal components PHQd and PHId from the sum quadrature and
inphase signals Qs and Is respectively, to generate modified
quadrature and inphase signals Qsc and Isc free of PRBS components
respectively.
[0076] At block 524, the method includes estimating quadrature and
inphase sum filter responses HQs and HIs affecting corresponding
PRBS signal components PHQs and PHIs in the difference quadrature
and inphase signals Qd and Id respectively.
[0077] At block 526, the method includes estimating an IQ mismatch
compensation filter response HIc(f) based on the estimated
quadrature and inphase sum filter responses HQs and HIs.
[0078] At block 528, the method includes filtering the modified
inphase signal Isc with the IQ mismatch compensation filter
response HIc(f), to generate a filter compensated inphase signal
ImHQs.
[0079] At block 530, the method includes applying mixer impairment
calibration and compensation on the filter compensated quadrature
and inphase signals Qfc and Ifc, to generate mixer compensated
quadrature and inphase signals Qmc and Imc respectively.
[0080] FIG. 6 is a schematic block diagram of a DDC front end
receiver 600, in accordance with an embodiment of the present
disclosure.
[0081] The DDC front end receiver 600 includes an antenna 602 for
receiving an input radio frequency (RF) signal s(t), and a signal
adder 604 for adding a RF PRBS signal p(t) to the input RF signal
s(t), to generate a modified input signal (r(t)=s(t)+p(t)).
[0082] The DDC front end receiver 600 further includes a
super-heterodyne upconverter 606 to up-convert a baseband PRBS
signal pbb(n) generated by a baseband PRBS generator 608, to the RF
PRBS signal p(t).
[0083] The DDC front end receiver 600 further includes first and
second mixers 610a and 610b for receiving the modified input signal
(r(t)=p(t)+s(t)) and generating modified baseband Inphase (Im) and
Quadrature (Qm) signals respectively, and first and second LPFs
612a and 612b for filtering the modified baseband Inphase (Im) and
Quadrature (Qm) signals, to generate filtered Inphase (If) and
Quadrature (Qf) signals respectively. There is a mismatch between
the modified baseband Quadrature (Qm) and Inphase (Im) signals due
to mixer impairment, and there is a further mismatch between
filtered Quadrature (Qf) and Inphase (Qf) signals due to filter
mismatch. In addition to the input RF signal s(t), the injected
PRBS signal (p(t)) also undergoes IQ impairment.
[0084] ADCs 614 and 616 receive the filtered Quadrature (Qf) and
Inphase (If) signals and generate digital versions of corresponding
signals. An adder 618 generates an overall down-converted BB signal
r.sub.bb(n) from real Quadrature (Qf(n)) and Inphase (If(n))
signals. The overall down-converted baseband signal r.sub.bb(n) is
represented by the following expression:
r.sub.bb(n)=k.sub.sig(n)*s.sub.bb(n)+k.sub.img(n)*s.sub.bb*(n)+k.sub.sig-
(n)*p.sub.bb(n)+k.sub.img(n)*p.sub.bb*(n) (26)
[0085] Where, [0086] k.sub.sig(n)=signal transfer function [0087]
k.sub.img(n)=image transfer function, both k.sub.sig(n) and
k.sub.img(n) defined as per equation 3, based on relative filter
imbalance (HQ(f)/HI(f)), and relative gain and phase imbalance of
the first and second mixers 610a and 610b. [0088]
s.sub.bb(n)=baseband input signal [0089] s.sub.bb*(n)=baseband
input image signal [0090] p.sub.bb(n)=baseband PRBS signal [0091]
p.sub.bb*(n)=baseband PRBS image signal
[0092] As seen in equation (26), apart from the desired baseband
input signal (s.sub.bb(n)), a scaled version of an undesired
baseband input image signal (s.sub.bb(n))*, a scaled version of the
PRBS signal (p.sub.bb(n)), and a scaled version of the undesirable
PRBS image signal (p.sub.bb*(n)) also appears in the overall
down-converted baseband signal r.sub.bb(n).
[0093] An IQ mismatch calibration block 620 includes a PRBS filter
based signal transfer function k.sub.sig(n) estimation block 620a,
and a PRBS filter based image transfer function k.sub.img((n)
estimation block 620b. In one embodiment, each of the blocks 620a
and 620b employ the PRBS based filter estimation block 300 (For
e.g., see FIG. 3) to estimate the signal transfer function
k.sub.sig(n) affecting the PRBS signal p.sub.bb(n), and image
transfer function k.sub.img(n) affecting the PRBS signal
p*.sub.bb(n) in presence of the down-converted BB signal
r.sub.bb(n). In one embodiment, each of the blocks 620a and 620b
receives the down-converted signal r.sub.bb(n) from the adder 618,
and the baseband PRBS signal p.sub.bb(n) from the baseband PRBS
signal generator 608.
[0094] The PRBS filter based signal transfer function estimation
block 620a essentially does the match filtering. On match
filtering, the signal and image component of the original BB signal
undergoes a processing gain of (1/M) in the above expression,
accounting for the estimation error in k.sub.sig(t). Filtering
r.sub.bb(n) through matched filter w.r.t p*.sub.bb(n) (i.e)
p.sub.bb(-n) gives
r bb ( n ) * p bb ( - n ) = { k sig ( n ) * s bb ( n ) * p bb ( - n
) + k img ( n ) * s bb * ( n ) * p bb ( - n ) + k sig ( n ) * p bb
( n ) * p bb ( - n ) + k img ( n ) * p bb * ( n ) * p bb ( - n ) r
bb ( n ) * p bb ( - n ) = { k sig ( n ) * s bb ( n ) * p bb ( - n )
+ k img ( n ) * s bb * ( n ) * p bb ( - n ) + k img ( n ) * R bb (
n ) r bb ( n ) * p bb ( - n ) = k ^ img ( n ) = k img ( n ) +
.delta. k img ( n ) ##EQU00006##
[0095] The PRBS based filter estimation block 620b also does the
match filtering. Filtering r.sub.bb(n) through matched filter w.r.t
p.sub.bb(n) (i.e) p*.sub.bb(-n) gives
r bb ( n ) * p bb * ( - n ) = { k sig ( n ) * s bb ( n ) * p bb * (
- n ) + k img ( n ) * s bb * ( n ) * p bb * ( - n ) + k sig ( n ) *
p bb ( n ) * p bb * ( - n ) + k img ( n ) * p bb * ( n ) * p bb * (
- n ) r bb ( n ) * p bb * ( - n ) = { k sig ( n ) * s bb ( n ) * p
bb * ( - n ) + k img ( n ) * s bb * ( n ) * p bb * ( - n ) + k sig
( n ) * R bb ( n ) r bb ( n ) * p bb * ( - n ) = k ^ sig ( n ) = k
sig ( n ) + .delta. k sig ( n ) ##EQU00007##
[0096] An IQ mismatch compensation block 622 uses the signal
transfer function k.sub.sig(n) and image transfer function
k.sub.img(n) estimated by the IQ mismatch calibration block 620,
for compensating the IQ impairment in the overall down-converted BB
signal r.sub.bb(n), and generates an IQ compensated baseband signal
r.sub.bb,comp(n), which is represented by the following
expression:
r.sub.bb,comp(n)=k.sub.sig*(n)*r.sub.bb(n)-k.sub.img(n)*r.sub.bb*(n)
(27)
[0097] Combining equation 26 and 27,
r bb , comp ( n ) = { ( k sig * k sig * ( n ) - k img * k img * ) *
S bb ( n ) + ( k sig * k sig * ( n ) - k img * k img * ) * P bb ( n
) ( 28 ) ##EQU00008##
[0098] The IQ mismatch compensation block 622 removes the image
component and the PRBS component from the overall down-converted BB
signal r.sub.bb(n) and generates the IQ compensated baseband signal
r.sub.bb,comp(n). However, the IQ compensated baseband signal
r.sub.bb,comp(n) has the K.sub.sig*K.sub.sig*P.sub.bb component
which may degrade noise.
[0099] A PRBS cancellation block 624 receives the baseband PRBS
signal p.sub.bb(n) and subtracts the residual PRBS signal
(k.sub.Sig(n)*p.sub.bb(n) and k.sub.img(n)*(p.sub.bb*(n)) from the
IQ compensated baseband signal r.sub.bb,comp(n) to generate a IQ
compensated baseband input signal s.sub.bbcomp(n), that is free of
PRBS signal.
[0100] FIG. 7 is a schematic block diagram of a signal generation
block 700 for generating a PRBS signal of RF, in accordance with an
embodiment of the present disclosure. In one embodiment, the signal
generation block 700 is placed between the baseband PRBS signal
generator 608, and the super heterodyne upconverter 606 to generate
the RF PRBS signal, free of any IQ mismatch. The signal generation
block 700 includes first and second zero-order hold image rejection
filters 702 and 704 for receiving and filtering baseband Quadrature
pQ.sub.bb(n) and Inphase pI.sub.bb(n) PRBS signals, first and
second mixers 706 and 708 for up-converting the filtered Quadrature
pQ.sub.bb(n) and Inphase pI.sub.bb(n) PRBS signals to an
Intermediate Frequency (IF), in discrete domain. The RF PRBS signal
generation block 700 further includes a subtractor 710 for
generating a difference of the upconverted Quadrature pQ.sub.IF(n)
and Inphase pI.sub.IF(n)
[0101] DAC 712 converts the digital difference signal to analog
signal, and a third real mixer 714 further upconverts the PRBS
signal to the Radio frequency. The third real mixer 714 is an all
digital mixer with IF=Fs/4 and is implemented as an adder. This
needs the input to be held for 4 clocks. This results in "hold
images" which needs to be filtered, further digitally and which
generates PRBS at IF without any IQ imbalance. The intermediate
Frequency (IF) is upconverted by the third real mixer 714 to the
desired RF band. The RF output is gained sufficiently to adequate
level before being added to the RF input of the receiver.
[0102] FIG. 8 is a flow diagram chart of a method of IQ mismatch
calibration and compensation in the DDC front end receiver, in
accordance with an embodiment of the present disclosure. In certain
embodiments, operations of method 800 are performed in the DDC
front end receiver, such as, for example, the DDC front end
receiver 600 (see, e.g., FIG. 6).
[0103] At block 802, the method includes receiving an input radio
frequency (RF) signal s(t).
[0104] At block 804, the method includes adding a pseudo random
binary sequence (PRBS) signal p(t) of Radio frequency (RF), to the
input RF signal s(t) to generate a modified input RF signal
(r(t)=s(t)+p(t)).
[0105] At block 806, the method includes mixing the modified input
RF signal (r(t)=s(t)+p(t)) with first and second orthogonal
sinusoid signals for generating baseband quadrature (Qm) and
inphase (Im) signals respectively.
[0106] At block 808, the method includes filtering the baseband
quadrature and inphase signals with quadrature and inphase filter
responses HQ(f) and HI(f), for generating filtered quadrature Qf
and inphase If signals respectively;
[0107] At block 810, the method includes converting the filtered
quadrature Qf and inphase If signals to a digital form.
[0108] At block 812, the method includes adding the digitized
filtered quadrature and inphase signals, to generate an overall
down-converted baseband signal rbb(n), wherein the overall
down-converted baseband signal includes baseband input and PRBS
signals filtered by a signal transfer function k.sub.sig, and the
baseband input and PRBS image signals filtered by an image transfer
function k.sub.img.
[0109] At block 814, the method includes estimating the signal
transfer k.sub.sig and image transfer k.sub.img functions affecting
the baseband PRBS signal p.sub.bb(n) and image pbb*(n)
respectively, in the presence of the overall down-converted
baseband signal r.sub.bb(n), based on PRBS based filter the overall
down-converted baseband signal and the baseband PRBS signal.
[0110] At block 816, the method includes generating an IQ
compensated baseband signal r.sub.bb,comp(n) based on the overall
down-converted baseband signal and the estimated signal transfer
and image transfer functions.
[0111] At block 818, the method includes cancelling PRBS components
from the IQ compensated baseband signal r.sub.bb,comp(n) based on
the baseband PRBS signal and the estimated signal transfer and
image transfer functions, to generate an IQ compensated baseband
input signal sbb,.sub.comp(n).
[0112] Without in any way limiting the scope, interpretation, or
application of the claims appearing below, advantages of one or
more of the example embodiments disclosed herein include IQ
mismatch calibration and compensation with an IRR greater than 90
dB and calibration time less than 500 ms. Various embodiments of
the present disclosure provide a dual channel direct
down-conversion front end receiver, in which the dual channel
architecture enables generation of low power differential filtered
signals (ImHId) and (QmHQd), and estimating filter responses based
on the low power differential filtered signals result in a
decreased calibration time. The minimum number of ADC samples
(NSampPG) required for a processing gain of 70 dBc is estimated to
be 10.sup.7, which is a drastic reduction as compared to 10.sup.11
samples required in existing single channel PRBS based
architecture. At a sampling rate 250 MSPS of ADC, this translates
to a calibration time of 40 ms, which is a significant improvement
over a calibration time of 400 s in existing single channel PRBS
based architecture. The dual channel DDC front end receiver is not
only limited to a Multi carrier GSM (MCGSM) base station, and can
be adapted for other radio standards as well.
[0113] It should be noted that reference throughout this
specification to features, advantages, or similar language does not
imply that all of the features and advantages should be, or are in,
any single embodiment. Rather, language referring to the features
and advantages is understood to mean that a specific feature,
advantage, or characteristic described in connection with an
embodiment is included in at least one embodiment of the present
disclosure. Thus, discussions of the features and advantages, and
similar language, throughout this specification do not necessarily,
refer to the same embodiment.
[0114] Various embodiments of the present disclosure, as discussed
above, are practiced with steps and/or operations in a different
order, and/or with hardware elements in configurations which are
different than those which are disclosed. Therefore, although the
disclosure has been described based upon these example embodiments,
it is noted that certain modifications, variations, and alternative
constructions are apparent and well within the spirit and scope of
the technology. Although various example embodiments of the present
disclosure are described herein in a language specific to
structural features and/or methodological acts, the subject matter
defined in the appended claims is not necessarily limited to the
specific features or acts described above. Rather, the specific
features and acts described above are disclosed as example forms of
implementing the claims.
* * * * *